SOCIAL SIMULATION SYSTEM WITH PROFILING MEANS APPLIED
`
`DOCUMENT ID DATE PUBLISHED
`
`JP 2006146858 A
`
`2006-06-08
`
`INVENTOR INFORMATION
`
`CITY
`
`N/A
`
`STATE
`
`N/A
`
`ZIP CODE
`
`COUNTRY
`
`N/A
`
`N/A
`
`NAME
`
`SHIMA HIROKI
`
`DATE FILED
`
`2004-11-23
`
`KWIC Hits
`
`Abstract
`
`<P>PROBLEM TO BE SOLVED: Toprovide a social simulation system for predicting or calculating
`social or market dynamics that can construct a practical application suppressing divergence between a
`simulated world and the real world, and returns calculations obtained in the simulation system or
`information obtained in the calculation process to the real world in a form profitable for activities of
`individuals in the real world who have cooperated in the simulation or activities of arbitrary social
`entities related to the individuals. <P>SOLUTION: Personal information collected by a concrete
`system for automatically collecting personal information in the real world is used for defining agents in
`a simulated world to reflect personal behavioral principles in the real world to the simulated world, so
`that a practical application suppressing divergence from the real world can be constructed.
`<P>COPYRIGHT. (C)2006, JPO&NCIPI
`
`Description
`
`Social simulation system using profiling means
`
`TECHNICAL-FIELD
`
`The present invention relates to a social simulation system for predicting or calculating social or
`market movements.
`
`BACKGROUND-ART
`
`Here, the prior art is organized from the three aspects related to the present invention: social
`simulation technology, technology for utilizing personal information, and technologyfor collecting
`personal information.
`
`Social simulation predicts or calculates social and market movements. Social simulations include
`those using techniques such as macro models, micro models, system dynamics, cell automata, and
`agent-basedsocial simulations. Among them, the present invention targets a plurality of agents and
`modules that have internal states and interact with each other in the simulation world, corresponding
`to social entities such as individuals and organizations in the real world. It is a social simulation
`characterized by calculating each agent and module and the overall state change with progress, and
`includes those using techniques such as cellular automata and agent-based social simulation. Utilizing
`technology that enables reliable social simulation in a situation whereit is difficult to predict the effects
`of policies and marketing strategies due to the diversification and complexity of social and market
`
`

`

`structures Doing so Is very important in making good decisions. However, even though the most
`traditional calculation methods using relatively simple statistical methods using macro models can
`handle simple social structures that can be expressed by typical functions, However, there is a serious
`limitation that social phenomenathat progress in a diverse and complex social structure cannot be
`handled in an appropriate form.
`
`In contrast, the system dynamics method, which analyzes phenomena as causal relationships,
`systematically captures the inputs and outputs of system components, describes the causal
`relationships using difference equations, and performs a simulation that traces the passage of time. By
`doing so, more complex social structures can be handled as simulation targets, so it is possible to
`handle social phenomena that could not be handled by the calculation method using the macro model.
`For example, the WORLD model, well-known asa social simulation using system dynamics, presents
`shocking calculation results to inform the world of the seriousness of global environmental problems
`and resource problems, and for subsequentpolicy formation and social development. This greatly
`contributed (for example, Non-Patent Document 1). However, this method hasa limitation that it
`cannot handle a more complicated social structure involving interactions between individuals
`constituting a society or a phenomenon causedbythe diversity of individuals.
`
`In contrast, social simulations using cell automata and agent-based model methods can handle
`interactions between individuals, individual diversity, changesin individual internal state, etc. as
`computation targets. The model has a high degree of freedom and can handle more complex social
`structures. In particular, agent-based social simulations can handle individual diversity, the ability to
`handle dynamic relationships between individuals, and can examine both individual and overall
`movements. From the advantage of being able to handle complex logical structures, it has been
`expected that more accurate prediction and analysis can be performedthan relatively simple
`calculation methods using spreadsheet software. . However, in simulation systems using conventional
`social simulation technology, including agent-based social simulation methods, models are abstract or
`even if they are concrete, real-world phenomena and simulation results Thereis still a problem thatit
`is difficult to construct a practical application because the divergenceis too large. For example, there
`are still no successful examples of building practical applications for specific purposes such as the
`formulation of corporate marketing strategies. Many of the social simulations that have been
`constructed or proposed to date have been used for theoretical research in the social sciencefield. It
`is not necessarily fatal that the general correspondenceis unclear (for example, Non-Patent Document
`2). However, this is a fatal problem whentrying to build an application. Therefore, developing a
`technology that suppresses the discrepancy betweenthe real world and the simulation world is a
`major issue in the technical field.
`
`As a measureto improvethis situation, we conducted questionnaires using paper media, web pages,
`mobile phones, etc. for individuals living in the real world, and reflected the collected personal
`information in the simulation world. A social simulation method has been proposed in which the
`behaviorprinciple of a social subject is reflected in the behavior principle of an agent in the simulation
`world to suppress the deviation from the real world (for example, Non-Patent Document 3). However,
`with this method, it is not always possible to obtain accurate and detailed personal information
`necessary for conducting an effective social simulation in order to construct a practical application.
`Because, when conducting a questionnaire, by providing some form of compensation to the
`respondents of the questionnaire, even if the incentive to answeritself can be used, the incentive to
`answer correctly should be usedIt is becauseit is not possible. Therefore, the problem that the
`divergence between the simulation world and the real world cannot be madesufficiently small
`remains. In addition, since personal information was processedinto information thatlost
`correspondencewith the individual and applied to the simulation world, the information obtained by the
`simulation was transferred to the individual who provided the personal information and those who were
`involved with the individual. As a result, the use of simulations waslimited.
`
`On the other hand, methodsfor utilizing personal information industrially have already been
`performed in various forms. One of the mosttraditional methods of analyzing social structure and
`market structure using personal information is simple tabulation or cross tabulation (aggregating each
`item with an axis to clarify the relationship between multiple question items) ), Multi-dimensional cross
`
`

`

`tabulation (things that perform cross tabulation in multi-dimensions), hierarchical tabulation
`(summarizing items such as “large sum’, “medium plan’, “subtotal” by another item) , Group tabulation
`(grouping and summing up field items in groups, such as when "400 yen" or "455 yen" is in the 400
`yen range), etc.), integrated tabulation (integrating multiple tables) Using traditional social survey
`results analysis methods, relatively simple spreadsheet software, statistical processing languages, and
`other methods. However, this method haslimitations in terms of the complexity of the logical structure
`that can be constructed, and the interaction between social actors and changesin the internal state of
`social actors cannot be handled as computational materials as they are. It has been difficult to obtain
`sufficient accuracy for forecasting and analyzing complex dynamics of society and markets.
`
`One of the relatively new ways of using personal information is to use data mining technologythat
`finds meaningful correlations and patterns from a vast amount of personal information. In data mining,
`personal information such as product purchaseresults, credit card usage history, communication
`history, access history, questionnaire responses, etc. is used by using techniques such as knowledge
`engineering, neural network,fractal, pattern recognition, fuzzy theory, and statistical analysis. Extract
`meaningful correlations and patterns. In contrastto traditional statistical analysis methods that focus
`on hypothesis verification, data mining focuses on finding hypotheses. Information extracted by data
`mining includes, for example, risk management such as customer segmentation and target marketing
`such as product promotion and target narrowing by selection of potential customers, detection of
`unauthorized use of credit cards and calculation of automobile insurance premium calculation
`materials, It is used for etc. Data mining is a useful approach that makesit possible to process and
`effectively use a large amountof information that cannot be processed by the human brain. However,
`data mining analysis is only for individual static data, and it is not aimed at calculating the impact of
`social dynamics. Can not.
`
`Another method ofutilizing personal information that has become popular in recent years is CRM
`technology that manages personal information obtained through communication with customersin an
`appropriate manner and improves the efficiency of subsequent communication and service provision.
`Is used. This is because the contents of communications and transactions with customers who have a
`relationship with each other are recorded in the database each time, and a system that can be referred
`to during new communications and transactions Is built up. It is something to share with. As a result,
`even if the person in charge changes, the information once confirmed can be confirmed on the
`system, soit is not necessary to confirm the same information twice. For example, whena delivery
`service provider or taxi dispatcher receives a call from a customer who has madea transaction, the
`customer can refer to the customer information from the database using the information on the
`incoming call number and check the address and name. This eliminates the need for the customer to
`conveythis information again. By recording information such as the model of the product used by the
`customer who communicated once and the content of the communication, the manufacturerwill not
`explain the model to be used or past exchanges as long as the customer numberis given. You can do
`it. By recording the information of customers who have made transactions once, the dealer does not
`have to check information such as the delivery address and billing address each time. It is possible to
`propose new products based on information such as past transaction history. However, CRM is aimed
`at improving the efficiency of communication with customers and realizing effective marketing targeting
`customers, and it has not been assumedthat personal information managed by CRM will be used for
`social simulation. .
`
`In addition, as an industrial utilization method of personal information, personal information on the
`past behavior and the currentsituation of the individual is accumulated in a database and provided
`with means such as recording, searching, and referencing information. To identify or authenticate an
`individual by collecting information that identifies the individual and information about the property or
`information unique to the individual in a database and providing means for registration, reference, etc.
`Various methods such as recording personal information on personal preferences, past behavior,
`behavior patterns, etc. in a database, and customizing the services providedto the individual and the
`content of the information based on the personal information It is done.
`
`As described above, the methodof utilizing personal information industrially has been dramatically
`improved in recent years. However, the personal information of individuals who are active in the real
`
`

`

`world is not processed by a method suchastabulation, but is directly used as personal information for
`analysis, which involves direct or indirect interactions between social actors. The technology to
`calculate the process of social dynamics has not been realized in the past.
`
`On the other hand, the technology for collecting personal information has also made dramatic
`progress in recent years. First, there is a system for collecting a history of access to a personal
`Internet website. This uses technology called cookies to write identification information for identifying
`the accessor to the device on the client side, and records the accessor's access information identified
`by the identification information each time communication is performed. How it works. Detailed
`information such as when and what page was viewedbythe visitor, what information wasclicked,
`whatinformation was presented, what was entered, what was purchased, etc. automatically Can be
`recorded. In addition, DoubleClick Inc., a major U.S. advertising agency, identified how visitors
`performed on many websites byidentifying them using the same identification information across
`multiple websites. We have developed a technology to acquire information in an integrated manner.
`The collected information is mainly used for data mining and database marketing, and has not been
`assumedto be used for social simulation.
`
`Second, there is a system that manages personal information collected through the customer
`service, call center or contact center. This is a mechanism in which a person in charge records
`information corresponding to the contents of communication with a customer through direct dialogue,
`telephone, e-mail, etc. as digital information. The collected information is used mainly for facilitating
`communication with customers and product proposals by analyzing customer preferences and
`purchasing trends, and has not been assumedto be used for social simulation.
`
`Third, there is a system that records personal information such as the contents of services used,
`provided information, and input contents while logged in to an account. This is a system that provides
`an accountto an individual and provides some service or information to the individual who is loggedin
`to the account by entering the ID and password issuedto the individual. In the process of using the
`service or referring to the information in a state in which the information is entered, the information that
`the individual inputs to the system, the information that the system presents to the individual, the
`information that identifies the information, or It has a function of recording information generated from
`such information in a database as personal information of the individual. The collected information is
`used for data mining, CRM, customization, etc. and has not been assumedto be used for social
`simulation.
`
`Fourthly, there is a system for recording personal information such as actions taken bythe individual
`and services provided to the individual in the process of using the card or IC by the individual. This is a
`system in which a card orIC is issued to an individual, and a service that provides some service or
`information to an individual who presents the card or IC issued to the individual, in which the individual
`presents the card or IC to the system In the process of using the service or referring to the information,
`the information on the action taken by the individual or the service provided to the individual is
`recorded in the database as the personal information of the individual. ICs and cards are being used
`as a means for individuals to use various services in the real world. Plastic magnetic cards have long
`been used for cash cards and credit cards issued byfinancial institutions and membership cards
`issued by companies that provide services for members. In recent years, cards equipped with ICs,
`such as Felica developed by Sony Corporation, have become widespread. The IC enables a
`recordable memory capacity to be expanded, highly secure encryption, non-contact communication,
`and thelike, and greatly expands the functions that a conventional plastic card has assumed. Because
`of its small size, it can be embeddedin a mobile phone or embeddedin various things as an IC tag.
`These cards and ICs can record various personal information. It is also possible to record personal
`identification information for identifying an individual and to extract detailed personal information
`corresponding to the personal identification information from the remote personal information
`managementserver using the personal identification information. Among personal information
`managedin such a form, there is a highly reliable one that has undergone a process such as being
`confirmed by an identification card at the time of registration, or updated in real time when the person
`uses the IC or the card. This includes the latest additions and the exact usages recorded. Such
`personal information is used to provide services that can be used by holding ICs and cards by users,
`
`

`

`as well as being used for inventory and sales managementas in the POS system, and data mining. It
`is often used for grasping the best-selling products and consumption trends according to the
`technique, and for target marketing such as providing advertisements and product information
`according to individual attributes. In these forms ofuse, it is already done to predict or calculate social
`or market movements. However, these are usually done using traditional aggregation methods, special
`search methods, or special algorithms such as neural computing, and data, programs that reflect
`personal information, or A social simulation system that performs prediction or calculation by
`dynamically controlling both has not been realized.
`
`Fifth, there is a method of acquiring personal information in an agent program that supports or
`represents personal decision making and actions. This is because an agent program that supports or
`represents a person's decision-making and actions inputs personal information about the person's
`policy, current state, past action history, etc. from the person, a person other than the person, or
`another device. Is provided, and the personal information acquired by the means is used to determine
`an action for supporting or acting on behalf of the individual's decision making or action. For example,
`there are practical examples such as an agent program that collects information on behalf of the
`person and an agent program that detects financial conditions on behalf of the person and performs
`financial transactions. In these agent programs, accurate and detailed personal information including
`the principals of the principal is input by the user of the agent program as necessary. This is becauseit
`is necessary for the user to input accurate and detailed information in order to use the agent program
`as desired. For example, an agent program that collects information is input to the person's interests,
`that is, what information is requested, and the agent program that performs financial transactions is
`the principal. Information on the behavioral principle of the financial transaction, for example, how
`much the marketprice of the financial asset held by the companyis sold. These are personal
`information with higher accuracy than personal information that is usually input by a method such as a
`questionnaire. However, until now, agents in agent-based social simulations have not been effectively
`linked with agent programs that support or act on behalf of individuals.
`
`[1] Donella H. Meadows, etal. “Limits of Growth Rome Club* The Crisis of Mankind "Report, Diamond,
`1979 [2] Kiyoshi Izumi, “Artificial Market-A Complex Approach to Market Analysis”, Morikita Publishing,
`July 2003 [3] Hiroki Shima "Application of Profiling Technology-Technology that Supports Diversified
`and Complex Society and People's Life" (FIF Special Report No. 9) Fujita Institute for Future
`Management, March 2004
`
`DISCLOSURE
`
`TECH-PROBLEM
`
`The main problems to be solved by the present invention are the following two points. First, ina
`social simulation system for predicting or calculating social or market dynamics, it is possible to
`suppress the divergence between the simulation world and the real world and to construct practical
`applications. Second, the calculation results obtained in the simulation system or the information
`obtained in the calculation process are useful for the activities of individuals in the real world who
`cooperated with the execution of the simulation or the activities of any social actors involved with the
`individuals. In form, it is to return to the real world.
`
`Social simulations constructed or proposed to date have a high level of abstraction in the model, and
`the correspondence between real-world social actors and things and simulation-world agents and
`modulesis unclear or insufficient. The existence and behaviorof social entities in the real world, or the
`relationship between social actors, at each level, the existence and behavior of agents in the
`simulation world, or the relationship between agents. And did not match or approximate.
`
`In particular, one of the reasons whythereliability of social simulation predictions cannot be obtained
`is that the simulation results greatly depend on the parameter settings in the social simulation.
`However, in the conventional social simulation, there is no definitive method for setting an accurate
`parameter. For example, in the conventional simulation, most of the parameters related to the
`
`

`

`preference and preference of agents in the simulation world are determined by random numbers. |
`haveto.
`
`In a social simulation executed on the premiseof the dissociated state, expect that the phenomenain
`the real world and the phenomena in the simulation world match or approximateat least at the
`individual level, and in many casesat the overall level.
`| can't. Therefore, it has been impossible to
`realize a practical application on the assumption that the coincidence or approximation is expected.
`
`TECH-SOLUTION
`
`In the present invention, by defining an agent in the simulation world using personal information
`handled by a specific system that collects or manages personal information of the real world individual,
`the principle of individual behaviorin the real world is reflected in the simulation world. This makesit
`possible to construct a practical application that suppresses the deviation from the real world.
`
`In the social simulation system or the method of executing the social simulation according to claim 1,
`the social simulation server receives personal information of an individual who is active in the real
`world from a personal information collection managementclient, and simulatesit. This is reflected in
`agents operating in the world (see FIGS. 1 and 2).
`
`According to the social simulation system or the method of executing the social simulation according
`to claim 5, the personal information management server mediates communication between the social
`simulation server and the personal information collection managementclient, so that the efficiency of
`personal information is improved. (See FIG. 3).
`
`In the social simulation system or the method of executing the social simulation according to claim 6,
`the social simulation server requests the personal information managementserver or the personal
`information collection managementclient for personal information necessary for executing the social
`simulation. By providing such a function, the acquisition of personal information used for conducting
`social simulation is facilitated (see FIG. 4).
`
`In the social simulation system or the method for executing the social simulation according to claim 7,
`the personal information collection managementclient or personal information managementserver
`cannot encrypt or partially identify the personal information transmitted to the social simulation server.
`By protecting the personal information provider into privacy, the privacy of the personal information
`provideris protected (see FIG. 5).
`
`The social simulation system or the method for executing the social simulation according to claim 8,
`wherein the personal information transmitted from the personal information collection management
`client to the social simulation serveris desirable for the social simulation practitioner or the personal
`information provider. (See FIG. 6).
`
`10. The social simulation system or method for executing a social simulation according to claim 9,
`wherein the personal information collection managementclient is an agent program that functions
`autonomously and has a function of supporting or acting as a personal decision or action. Is used.
`
`The social simulation system or the method of executing the social simulation according to claim 10,
`wherein the personal information collection and managementclient is an IC scanner or card having a
`function of reading personal information from an IC or card on which personal information is recorded.
`You are using a scanner.
`
`In the social simulation system or the method of executing the social simulation according to claim
`11, as the personal information collection managementclient, personal information reading means for
`reading personal information from a questionnaire response content recording medium such as paper,
`the personal information reading Personal information recording means for recording the personal
`information read by the meansasdigital information, and personal information transmitting means for
`reading the personal information recorded in the personal information and directly or indirectly
`transmitting it to the social simulation server. Uses the questionnaire / scanner provided.
`
`

`

`The social simulation system or the method for executing a social simulation according to claim 12,
`wherein a personal information collection and managementclient, as a personal information collection
`managementclient, arranges a question, a computer having a human or intelligent judgment
`mechanism, and a survey target answering the question A function to automatically record the
`contents of interview responsesin an interview processinvolving interaction with a person as digital
`information, or a social simulation server, a personal information managementserver, or a surveyed
`person already in the company. The personal information is referred to, and the information items that
`are further required for executing the social simulation are determined, the questioneris instructed
`about what to ask, and the answer content that the questioner has drawn from the respondent,
`Automatically or according to the questioner's input , The ability to record a digital information, having
`utilizes interviewee recording apparatus.
`
`14. The social simulation system or method for executing a social simulation according to claim 13,
`wherein the navigation has a function of guiding a route of a user or a transportation means used by
`the user in an arbitrary environment as a personal information collection managementclient. -A
`system or navigation device is used.
`
`In the social simulation system or the method of executing the social simulation according to claim
`14, any software having a main function other than the function of collecting personal information for
`use in the social simulation is used as a personal information collection managementclient. We are
`using.
`
`In the social simulation system or the method for executing the social simulation according to claim
`15, as a personal information collection managementclient, means for observing the surrounding
`situation in an arbitrary environment, means for converting the observation contents into digital
`information, And a sensor or a measuring device including the sensor, which includes one or more of a
`recording unit that records the digital information and a transmission unit that transmits the digital
`information to another device.
`
`The social simulation system or method for executing a social simulation according to claim 16,
`wherein in the personal information collection and managementclient, means for reading personal
`identification information from personal identification information recording means for recording
`personal identification information; Means for transmitting information to a remote personal information
`management server network-connected to the personal information collection managementclient
`having a function of managing personal information, and the personal identification information
`transmitted from the remote personal information management server Means for receiving
`corresponding personal information, and in the remote personal information management server,
`meansfor receiving the personal identification information from the personal information collection and
`managementclient, and the personal information from a database held or accessible by itself Means
`for extracting personal information corresponding to the information identification information, and the
`personal information By and meansfor transmitting the original personal information collection
`managementclient, thereby enabling the use of detailed personal information managed by the
`network.
`
`The social simulation system or the method for executing the social simulation according to claim 17,
`wherein the personal information collection managementclient uses information input by a personal
`information provider to a device for browsing or viewing digital information, Means for manipulating the
`information presented to the individual through the identification information identifying the information
`or elements closely related to the information embeddedin the information presented to the individual
`through the device; Or recording personal information related to an act of browsing or viewing digital
`information of an individual using the device, using information input by the individual through the
`device in responseto the information, or both, or Ahuman digital information behavior monitoring
`system comprising means for transmitting to another device network-connected to the deviceIt is
`using.
`
`In the social simulation system or the method of executing the social simulation according to claim
`18, by applying an agent-based social simulation as a simulation method, it is possible to handle a
`
`

`

`social structure similar to the real world in the simulation world.
`
`It is said.
`
`The social simulation system or method for executing a social simulation according to claim 19,
`wherein the personal information collection and managementclient specifies personal information
`related to a product or service purchased or used by a personal information provider, and identifies the
`product or service. Means for recording using product service identification information that can be
`performed, and means for directly or indirectly transmitting the personal information to a social
`simulation server or personal information managementserver, the social simulation server or the In
`the personal information managementserver, the means for receiving the personal information, and
`the product service identification information for specifying the product or service purchased or used
`by the personal information provider constituting the personal information, the remote product service
`information management server Means for transmitting to the remote commodity service Means for
`receiving information indicating the detailed content of the product or service corresponding to the
`product service identification information transmitted from the service information managementserver,
`wherein the social simulation server or the Means for receiving the merchandise service identification
`information transmitted from the personal information management server; merchandise service
`identification information held by the personal information managementserver or accessible byitself:
`and detailed information on the corresponding merchandise or service. Means for extracting detailed
`information of the product or service corresponding to the